Metraipolis › archive › dc_9fb99356ee2d
Calibrating Artificial Guilt: Neurally Grounded Reward Shaping for Prosocial Multi-Agent Reinforcement Learning — arXiv:2608.04663
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.04663
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T11:12:39.567Z
- SHA-256 of the archived text
d7cb15d7041896e81ddef03629c997ac1c018262594a8f16e5f9e3400936c60a- Stable URL of this page
https://metraipolis.com/documento/dc_9fb99356ee2d
How to cite
Calibrating Artificial Guilt: Neurally Grounded Reward Shaping for Prosocial Multi-Agent Reinforcement Learning — arXiv:2608.04663 — arXiv (en), https://arxiv.org/abs/2608.04663 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9fb99356ee2d (Metraipolis archive dc_9fb99356ee2d, sha256 d7cb15d7041896e81ddef03629c997ac1c018262594a8f16e5f9e3400936c60a)
BibTeX
@misc{dc9fb99356ee2d,
title = {Calibrating Artificial Guilt: Neurally Grounded Reward Shaping for Prosocial Multi-Agent Reinforcement Learning — arXiv:2608.04663},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.04663},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9fb99356ee2d; sha256 d7cb15d7041896e81ddef03629c997ac1c018262594a8f16e5f9e3400936c60a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04663. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Cooperative multi-agent reinforcement learning often adds social terms to individual rewards, yet the scale of those terms is usually chosen by hand. We ask whether a guilt signal can instead be calibrated from human neural and behavioural data and transferred to artificial agents. Using the public SoDec responsibility fMRI dataset (40 participants), we fit a subject-fixed-effects regression of momentary-happiness changes on outcome-type counts and recover a guilt weight as the Partner-negative minus Social-negative contrast ($\hat{w}=1.118$, Cohen's $d=0.214$). We embed this weight in a two-agent Social Lottery environment and train independent Proximal Policy Optimization actor-critics under four shaping regimes: neurally calibrated, uniform constant, zero (selfish), and a unit-coefficient oracle. Across 1{,}000 evaluation episodes per condition, the calibrated agents track the human Social safe-choice rate most closely ($0.459$ vs.\ human $0.484$; $\mathrm{KL}=0.0012$), while the other three conditions deviate by one to three orders of magnitude in KL. Human neurobehavioural priors can therefore act as quantitative constraints on prosocial reward shaping.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-06 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.04663 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-31T11:02:13.420Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).